AICA Competition Banner

On-Site Finals Detailed Scenario

This page provides details of the onsite finals scenario that simulates a multimodal autonomous delivery system in an urban environment.



Scenario Summary

The On-Site Finals follow the Virtual Stage scenario, with more deliveries and buildings that take more than one package. Window deliveries pay a bonus that rises with height. One delivery window carries an additional priority bonus for early completion, representing an urgent delivery such as a medical emergency. The QCar2 is penalized for hitting the curb.

Success in this challenge requires developing effective delivery strategies and implementing efficient path-planning and control algorithms.

Figure 1: Overview of On-Site Finals cityscape scenario map


Scenario Setup

Delivery Information

  • The scenario includes 14 delivery tasks: 11 small-package deliveries and 3 large-package deliveries.
  • All packages are initially located at the central pickup location.
  • Two delivery options are available: shared drop-off and window delivery.
  • One out of 11 small-package deliveries is an urgent delivery. Its window pad is marked by orange pad. It pays an additional bonus if served within the first 2 minutes of the run.

On-Site Finals Files

  • The files for the On-Site Finals have been added to the GitHub repository. The new files are:
    game_finals.py
    setup_env_finals.py
    plot_trace.py
    driveable_grid.npz
    run_finals

Pickup and Delivery Locations

Figure 2: Central pickup and delivery locations for the On-Site Finals scenario.

Tables 1 list the pickup and delivery locations.

Delivery Location Table

Location Package Type Number
of Packages
Shared Drop-Off
Location [x y z]
Python
Node
Matlab
Node
Window Delivery
Location [x y z] (bonus)
Central pickup Pickup - [-2.50305 29.6703 0.05] 24 25 None
Building 1 Small 3 [11.2739 -10.84655 0.05] 2 3 Window 1 — [15.1739 -18.04655 9.65] (400)
Window 2 — [-3.492 -18.441 9.663] (400)
Window 3 — [3.475 -21.886 11.291] (400)
Building 2 Small 1 [22.435 1.392 0.005] 4 5 Window 4 — [27.787 0.358 4.291] (200)
Building 3 Small 2 [22.5478 29.6703 0.05] 14 15 Window 5 — [25.401 17.005 9.15] (300)
Window 6 — [25.32 35.156 13.321] (400)
Building 4 Small 2 [0.0 44.9735 0.05] 20 21 Window 7 — [1.3 47.5985 4.807] (100)
Window 8 — [-10.8 47.34 4.357] (100)
Building 5 Small 1 [-19.84125 29.6703 0.05] 22 23 None
Building 6 Small 1 None - - Window 9 — [10.804 27.047 4.265] (200)
Building 7 Small 1 None - - Window 10 — [-23.873 6.209 9.15] (300) — urgent delivery
Curbside 1 Large 1 [-12.8205 -4.5991 0.05] 10 11 None
Curbside 2 Large 1 [8.975 37.099 0.005] 16 17 None
Curbside 3 Large 1 [8.367 10.853 0.005] 6 7 None

Table 1: Pickup, shared drop-off and window delivery locations.

Figure 3: Road network with node numbering in Python.


Figure 4: Road network with node numbering in MATLAB/Simulink.


Scenario Rules

On-Site scenario rules follow the Virtual Stage scenario rules. The additions are summarized here.

  • When a pad becomes occupied, a box appears on it. An occupied pad accepts no further deliveries.
  • A window delivery occupies that window pad immediately.
  • A shared drop-off does not occupy the shared drop-off pad while the building still has deliveries outstanding.
  • Once a building's final delivery is made, all of its window pads and its shared drop-off pad become occupied, indicating that no further deliveries to that building are possible.
  • After 5 minutes, pads become occupied according to the number of deliveries completed.

Figure 5: All Deliveries Completed

Back to:

Onsite Finals Competition Guide

AICA Home Portal